WIP Prometheus without operator

bin/y-cluster-converge-ystack

@@ -71,17 +71,9 @@ apply_base 08-buildkitd-grpcroute
 k -n ystack get grpcroute buildkitd
 echo "# Validated: grpcroute buildkitd exists"
 
-# 7. Monitoring operator + CRDs
-apply_base 30-monitoring-operator
-echo "# Waiting for prometheus-operator CRDs to register ..."
-until k get crd prometheuses.monitoring.coreos.com >/dev/null 2>&1; do sleep 2; done
-until k get crd alertmanagers.monitoring.coreos.com >/dev/null 2>&1; do sleep 2; done
-until k get crd servicemonitors.monitoring.coreos.com >/dev/null 2>&1; do sleep 2; done
-echo "# Validated: prometheus-operator CRDs registered"
-
-# 8. Monitoring CRs (Prometheus, Alertmanager, exporters)
-apply_base 31-monitoring
-k -n monitoring get prometheus now
+# 7. Monitoring (vanilla Prometheus + Alertmanager + exporters)
+apply_base 30-monitoring
+k -n monitoring rollout status deploy/prometheus-now --timeout=120s
 echo "# Validated: monitoring stack exists"
 
 # 6.8 Prometheus HTTPRoute

bin/y-cluster-provision-k3d

@@ -129,5 +129,8 @@ PROD_REGISTRY_IP=$(kubectl --context=$CTX -n ystack get service prod-registry -o
 docker exec k3d-ystack-server-0 sh -cex "echo '$BUILDS_REGISTRY_IP builds-registry.ystack.svc.cluster.local' >> /etc/hosts"
 docker exec k3d-ystack-server-0 sh -cex "echo '$PROD_REGISTRY_IP   prod-registry.ystack.svc.cluster.local' >> /etc/hosts"
 
-echo "# Updating /etc/hosts (requires sudo) ..."
-y-k8s-ingress-hosts --context=$CTX -write -override-ip "${YSTACK_PORTS_IP:-127.0.0.1}"
+echo "# Checking /etc/hosts ..."
+if ! y-k8s-ingress-hosts --context=$CTX -check -override-ip "${YSTACK_PORTS_IP:-127.0.0.1}"; then
+  echo "# Updating /etc/hosts (requires sudo) ..."
+  y-k8s-ingress-hosts --context=$CTX -write -override-ip "${YSTACK_PORTS_IP:-127.0.0.1}"
+fi

bin/y-cluster-provision-lima

@@ -122,5 +122,8 @@ PROD_REGISTRY_IP=$(kubectl --context=$CTX -n ystack get service prod-registry -o
 limactl shell ystack sudo sh -c "echo '$BUILDS_REGISTRY_IP builds-registry.ystack.svc.cluster.local' >> /etc/hosts"
 limactl shell ystack sudo sh -c "echo '$PROD_REGISTRY_IP   prod-registry.ystack.svc.cluster.local' >> /etc/hosts"
 
-echo "# Updating /etc/hosts (requires sudo) ..."
-y-k8s-ingress-hosts --context=$CTX -write -override-ip 127.0.0.1
+echo "# Checking /etc/hosts ..."
+if ! y-k8s-ingress-hosts --context=$CTX -check -override-ip 127.0.0.1; then
+  echo "# Updating /etc/hosts (requires sudo) ..."
+  y-k8s-ingress-hosts --context=$CTX -write -override-ip 127.0.0.1
+fi

bin/y-cluster-validate-ystack

@@ -82,10 +82,10 @@ k -n ystack get grpcroute buildkitd >/dev/null 2>&1 \
   && report "grpcroute buildkitd" "ok" \
   || report "grpcroute buildkitd" "not found"
 
-# 7.6 Monitoring stack
-k -n monitoring get prometheus now >/dev/null 2>&1 \
-  && report "prometheus now" "ok" \
-  || report "prometheus now" "not found"
+# 7.6 Monitoring stack (vanilla Prometheus deployment)
+ROLLOUT_PROM=$(k -n monitoring rollout status deploy/prometheus-now --timeout=10s 2>&1) \
+  && report "prometheus-now rollout" "ok" \
+  || report "prometheus-now rollout" "$ROLLOUT_PROM"
 
 # 7.7 Prometheus HTTPRoute
 k -n monitoring get httproute prometheus-now >/dev/null 2>&1 \

k3s/30-monitoring/kustomization.yaml

@@ -0,0 +1,6 @@
+resources:
+- ../../monitoring/namespace
+- ../../monitoring/prometheus-now
+- ../../monitoring/alertmanager-main
+- ../../monitoring/kube-state-metrics-now
+- ../../monitoring/node-exporter-now

monitoring/alertmanager-main/main-alertmanager-service.yaml

@@ -9,4 +9,5 @@ spec:
     protocol: TCP
     targetPort: web
   selector:
-    alertmanager: main
+    app.kubernetes.io/name: alertmanager
+    app.kubernetes.io/instance: main

monitoring/alertmanager-main/main-alertmanager.yaml

@@ -1,6 +1,60 @@
-apiVersion: monitoring.coreos.com/v1
-kind: Alertmanager
+apiVersion: apps/v1
+kind: Deployment
 metadata:
-  name: main
+  name: alertmanager-main
+  labels:
+    app.kubernetes.io/name: alertmanager
+    app.kubernetes.io/instance: main
 spec:
   replicas: 1
+  selector:
+    matchLabels:
+      app.kubernetes.io/name: alertmanager
+      app.kubernetes.io/instance: main
+  template:
+    metadata:
+      labels:
+        app.kubernetes.io/name: alertmanager
+        app.kubernetes.io/instance: main
+    spec:
+      securityContext:
+        runAsUser: 65534
+        runAsGroup: 65534
+        runAsNonRoot: true
+        fsGroup: 65534
+      containers:
+        - name: alertmanager
+          image: quay.io/prometheus/alertmanager:v0.28.1
+          args:
+            - --config.file=/etc/alertmanager/alertmanager.yaml
+            - --storage.path=/data
+          ports:
+            - name: web
+              containerPort: 9093
+          readinessProbe:
+            httpGet:
+              path: /-/ready
+              port: web
+            initialDelaySeconds: 5
+          livenessProbe:
+            httpGet:
+              path: /-/healthy
+              port: web
+            initialDelaySeconds: 15
+          resources:
+            requests:
+              cpu: 10m
+              memory: 32Mi
+            limits:
+              memory: 64Mi
+          volumeMounts:
+            - name: config
+              mountPath: /etc/alertmanager
+            - name: data
+              mountPath: /data
+      volumes:
+        - name: config
+          secret:
+            secretName: alertmanager-main
+        - name: data
+          emptyDir: {}

monitoring/greptimedb/bucket-create.yaml

@@ -0,0 +1,35 @@
+apiVersion: batch/v1
+kind: Job
+metadata:
+  name: bucket-create-greptimedb
+spec:
+  template:
+    spec:
+      containers:
+      - name: mc
+        image: minio/mc:RELEASE.2025-08-13T08-35-41Z
+        env:
+        - name: AWS_ACCESS_KEY_ID
+          valueFrom:
+            secretKeyRef:
+              name: minio
+              key: accesskey
+        - name: AWS_SECRET_ACCESS_KEY
+          valueFrom:
+            secretKeyRef:
+              name: minio
+              key: secretkey
+        - name: BUCKET_NAME
+          value: greptimedb
+        - name: S3_ENDPOINT
+          value: http://blobs-versitygw.ystack.svc.cluster.local
+        command:
+        - sh
+        - -ce
+        - |
+          until mc alias set s3 $S3_ENDPOINT $AWS_ACCESS_KEY_ID $AWS_SECRET_ACCESS_KEY 2>/dev/null; do
+            sleep 2
+          done
+          mc mb --ignore-existing s3/$BUCKET_NAME
+      restartPolicy: Never
+  backoffLimit: 10

monitoring/greptimedb/config.toml

@@ -0,0 +1,7 @@
+[storage]
+type = "S3"
+bucket = "greptimedb"
+endpoint = "http://blobs-versitygw.ystack.svc.cluster.local"
+access_key_id = "YstackEXAMPLEKEY"
+secret_access_key = "github.com/Yolean/ystack-EXAMPLE"
+region = "us-east-1"

monitoring/greptimedb/greptimedb.yaml

@@ -0,0 +1,72 @@
+apiVersion: v1
+kind: Service
+metadata:
+  name: greptimedb
+spec:
+  ports:
+  - name: http
+    port: 4000
+    targetPort: http
+  - name: grpc
+    port: 4001
+    targetPort: grpc
+  selector:
+    app: greptimedb
+---
+apiVersion: apps/v1
+kind: Deployment
+metadata:
+  name: greptimedb
+  labels:
+    app: greptimedb
+spec:
+  replicas: 1
+  selector:
+    matchLabels:
+      app: greptimedb
+  template:
+    metadata:
+      labels:
+        app: greptimedb
+    spec:
+      containers:
+      - name: greptimedb
+        image: greptime/greptimedb:v0.12.0
+        args:
+        - standalone
+        - start
+        - --config-file=/etc/greptimedb/config.toml
+        - --http-addr=0.0.0.0:4000
+        - --rpc-addr=0.0.0.0:4001
+        ports:
+        - name: http
+          containerPort: 4000
+        - name: grpc
+          containerPort: 4001
+        readinessProbe:
+          httpGet:
+            path: /health
+            port: http
+          initialDelaySeconds: 10
+        livenessProbe:
+          httpGet:
+            path: /health
+            port: http
+          initialDelaySeconds: 15
+        resources:
+          requests:
+            cpu: 50m
+            memory: 256Mi
+          limits:
+            memory: 768Mi
+        volumeMounts:
+        - name: data
+          mountPath: /tmp/greptimedb
+        - name: config
+          mountPath: /etc/greptimedb
+      volumes:
+      - name: data
+        emptyDir: {}
+      - name: config
+        configMap:
+          name: greptimedb-config

monitoring/greptimedb/kustomization.yaml

@@ -0,0 +1,22 @@
+apiVersion: kustomize.config.k8s.io/v1beta1
+kind: Kustomization
+
+namespace: monitoring
+
+resources:
+- greptimedb.yaml
+- bucket-create.yaml
+
+generatorOptions:
+  disableNameSuffixHash: true
+
+secretGenerator:
+- name: minio
+  literals:
+  - accesskey=YstackEXAMPLEKEY
+  - secretkey=github.com/Yolean/ystack-EXAMPLE
+
+configMapGenerator:
+- name: greptimedb-config
+  files:
+  - config.toml

monitoring/kube-state-metrics-now/kustomization.yaml

@@ -5,12 +5,3 @@ namespace: monitoring
 
 resources:
 - ../kube-state-metrics
-
-patchesStrategicMerge:
-- |-
-  apiVersion: monitoring.coreos.com/v1
-  kind: ServiceMonitor
-  metadata:
-    name: kube-state-metrics
-    labels:
-      prometheus: now

monitoring/kube-state-metrics/kustomization.yaml

@@ -7,4 +7,3 @@ resources:
 - deployment.yaml
 - service-account.yaml
 - service.yaml
-- kube-state-metrics-servicemonitor.yaml

monitoring/node-exporter-now/kustomization.yaml

@@ -5,19 +5,3 @@ namespace: monitoring
 
 resources:
 - ../node-exporter
-
-patchesStrategicMerge:
-- |-
-  apiVersion: monitoring.coreos.com/v1
-  kind: PodMonitor
-  metadata:
-    name: node-exporter
-    labels:
-      prometheus: now
-- |-
-  apiVersion: monitoring.coreos.com/v1
-  kind: PrometheusRule
-  metadata:
-    name: node-exporter
-    labels:
-      prometheus: now

monitoring/node-exporter/kustomization.yaml

@@ -3,5 +3,3 @@ resources:
 - node-exporter-clusterRole.yaml
 - node-exporter-clusterRoleBinding.yaml
 - node-exporter-daemonset.yaml
-- node-exporter-podmonitor.yaml
-- example-rules.yaml

monitoring/prometheus-now/kustomization.yaml

@@ -7,3 +7,14 @@ resources:
 - ../rbac-prometheus
 - now-prometheus-service.yaml
 - now-prometheus.yaml
+
+generatorOptions:
+  disableNameSuffixHash: true
+
+configMapGenerator:
+- name: prometheus-now-config
+  files:
+  - prometheus.yml
+- name: prometheus-now-rules
+  files:
+  - rules/node-exporter.yml

monitoring/prometheus-now/now-prometheus-service.yaml

@@ -10,4 +10,4 @@ spec:
     targetPort: web
   selector:
     app.kubernetes.io/name: prometheus
-    prometheus: now
+    app.kubernetes.io/instance: now